Recursos

Dwarfing apple rootstocks may reduce scion vigour by restricting the uptake and transport of mineral nutrients to the scion. For example, composite trees of ‘McIntosh’ grown in solution culture and treated with 32P and 45Ca had greatly decreased transport of radioactivity from the root to the scion when grown on M.9 and compared with the M.16 rootstock (Bukovac et al., 1958). Jones (1971) also reported that concentrations of nitrogen, phosphorous and potassium in the xylem sap collected from root pressure exudate was reduced by M.7 compared with the MM.104 rootstock.

In contrast, Warne and Wallace (1935) measured the concentrations of mineral anions and cations in two scion varieties on a range of different size-controlling rootstocks and concluded that many beneficial attributes imposed by the M.9 rootstock, such as reduced scion vigour, increased floral precocity and fruit quality, could not be explained by differences in mineral composition within the leaves, bark, wood and fruit. Similarly,

the root system of M.9 grown in nutrient culture tended to absorb more water, nitrogen and potassium over the growing season than the M.12 root system when data for absorption by each rootstock was expressed as a proportion to the annual gain in fresh weight of the scion (Pearse, 1940). More recently, Atkinson and Else (2001) reported that the M.9 rootstock was actually capable of transporting greater amounts of mineral ions and solutes to the scion than MM.106 when differences in sap flow rate, root mass and leaf area imposed by each rootstock type were appropriately considered.

1.7.2 Altered plant water status

Dwarfing apple rootstocks may reduce scion vigour by restricting the movement of water from the root to the scion (Rogers and Beakbane, 1957; Olien and Lasko, 1986; Atkinson et al., 2003). Early anatomical studies reported that roots of dwarfing apple rootstocks had a reduced ratio of xylem to phloem and much smaller xylem vessels compared with roots of vigorous rootstocks (Beakbane and Thompson, 1947; Beakbane, 1956). Collectively, these anatomical differences of the root for dwarfing apple rootstocks were thought to limit axial water movement through the root system, hence restricting the transport of water to the scion.

In further support of this theory, the hydraulic conductivity of M.27, M.9 and MM.106 roots (1.5 mm in diameter) decreased with decreasing rootstock vigour, therefore indicating that the root system of dwarfing apple rootstocks may provide a greater resistance to axial water movement than rootstocks of greater vigour (Atkinson et al., 2003). However, the hydraulic conductivity of the entire M.9 root system was similar to the MM.106 root system when hydraulic conductivity was expressed relative to the total dry mass of the root system (Atkinson et al., 2003).

The increased axial resistance to water movement in the roots of M.9 (Atkinson et al., 2003) is unlikely to reduce water uptake by the root system, which is influenced solely by radial resistances to water movement (Steudle and Peterson, 1998). Arguably, some architectural differences in the M.9 root system might enhance the extraction of soil water when soil moisture is non-limiting. For example, a large proportion of the M.9 root system was comprised of roots less than 1 mm in diameter (Atkinson et al., 2003), which are typically produced on a seasonal basis (Eissenstat et al., 2001) and new white

roots are important for water and nutrient acquisition by apple trees (Atkinson, 1974; Eissenstat et al., 2001). Despite these probable benefits to water absorption, decreased axial water movement through the xylem of dwarfing root systems may explain why they sometimes decrease the stem water potential of the scion.

Olien and Lasko (1986) reported that M.9 and M.26 decreased the midday stem water potential of six-year-old ‘Empire’ scions compared with rootstocks of M.7, MM.106 and MM.104. Differences in stem water potential for M.9 and M.26 were not due to differences in stomatal conductance or transpiration but were attributed to reduced water flow through the rootstock portion of the tree (Olien and Lasko, 1986). Indeed, the two-year-old union of the ‘Bramley’ scion budded onto M.9 contained xylem vessels of reduced diameter and little xylem was produced in the rootstock tissue (Soumelidou et al., 1994b). In addition, the area of functional xylem at the bud union and in the above scion stem was reduced by M.27 and M.9 compared with MM.106, and this may have decreased the hydraulic conductance of the bud union for trees on M.27 and M.9 (Atkinson et al., 2003). Therefore, differences in graft union morphology for dwarfing apple rootstocks may impede water movement into the stem of the primary shoot, thereby explaining why the scion on dwarfing apple rootstocks developed more negative midday water potentials (i.e., Olien and Lasko, 1986).

However, dwarfing apple rootstocks are not consistently reported to decrease midday water potential of the scion. For example, changes in diurnal leaf water potential and stomatal conductance were almost identical when three-year-old ‘Queen Cox’ grafted onto M.9 and MM.106 rootstocks were grown without root-restriction and supplied with sufficient irrigation (Atkinson et al., 2000). It is important to note that in the study of Olien and Lasko (1986), decreased midday stem water potentials for the scion growing on M.9 or M.26 were, biologically, very small compared with rootstocks of greater vigour (i.e., -0.1 to -0.2 MPa). In addition, stem water potential of the scion in the canopy exterior did not exceed -1.0 MPa at midday for any rootstock type. For deficit irrigated apple trees, a midday leaf water potential greater than -1.5 MPa was required to significantly decrease shoot extension growth of the scion compared with fully irrigated controls (Irving and Drost, 1987), whereas photosynthesis was reduced as midday leaf water potential approached -1.8 to -2.5 MPa (Kilili et al., 1996; Mills et al., 1996). Arguably, relatively small reductions in stem water potential imposed by M.9 in

the study of Olien and Lasko (1986) may have been insufficient to decrease shoot extension growth and, therefore, are probably not the primary cause of scion dwarfing. Similarly, Webster (2004) reported that reductions in the hydraulic conductance at or above the graft union might not be physiologically sufficient to reduce shoot extension growth of the scion growing on dwarfing apple rootstocks.

In summary, dwarfing apple rootstocks do not consistently decrease midday water potential of the scion compared with rootstocks of greater vigour. Differences that occur amongst different size-controlling rootstocks are physiologically very small, and therefore are probably not the primary cause of rootstock-induced dwarfing of the scion.

1.7.3 Alterations in shoot/root/shoot transport of endogenous plant hormones

1.7.3.1 Effect of dwarfing apple rootstocks on shoot to root signalling of indole-3-acetic acid

Dwarfing apple rootstocks may reduce scion vigour by modifying the transport of endogenous hormones (Rogers and Beakbane, 1957; Lockard and Schneider, 1981; Webster, 1995; Kamboj and Quinlan, 1997; Atkinson and Else, 2001; Webster, 2004). The endogenous control of scion vigour by dwarfing apple rootstocks is most convincingly explained by hormonal signalling between endogenous indole-3-acetic acid (IAA) and cytokinin. Lockard and Schneider (1981) hypothesised that dwarfing apple rootstocks reduced scion vigour by decreasing the basipetal transport of IAA within the phloem and cambial cells of the rootstock stem, therefore limiting the amount of IAA transported from scion to root. Sub-optimal amounts of IAA transported to the root system may decrease root growth, the synthesis of root-produced cytokinins and their consequent transport in the xylem sap to the scion where cytokinins may be required to stimulate shoot extension growth (Lockard and Schneider, 1981). In agreement with this hypothesis, the stem of dwarfing apple rootstocks reduced the basipetal transport of radio-labelled IAA (Soumelidou et al., 1994a; Kamboj et al., 1997), whilst the concentration of endogenous IAA was decreased in their cambial sap (Michalczuk, 2002) when compared with rootstocks of greater vigour. In addition, composite trees of ‘Fiesta’ grafted onto M.9 transported less 3H to the root system

compared with rootstocks of greater vigour when 3H-IAA was applied to a mature basal leaf on the scion (Kamboj, 1996).

The mechanism by which the tissue of the rootstock stem decreases the basipetal transport of IAA is not known, but the bark of dwarfing apple rootstocks exhibited an increased capacity to destroy IAA (Gur and Samish, 1968) and contained higher concentrations of growth inhibiting phenols and lower concentrations of growth promoting phenols (Martin and Stahly, 1967) that may act to enhance or suppress the oxidative decarboxylation of IAA, respectively (Lockard and Schneider, 1981). In addition, dwarfing rootstocks may have an abnormal arrangement of efflux proteins that facilitate active polar IAA transport out of the cell (Soumelidou et al., 1994a; Kamboj, 1996; Kamboj et al., 1997). Although Kamboj (1996) showed that M.9 transported less 3

H to the root system compared with rootstocks of greater vigour when 3H-IAA was applied to a mature basal leaf on the scion, the proportion of 3H in the root system that was still associated with IAA was not determined for rootstocks of different vigour. Therefore, it remains untested experimentally whether the transport of ‘IAA’ to the root is actually decreased by M.9, and whether this is causal in decreased root growth of dwarfing apple rootstocks as proposed by Lockard and Schneider (1981).

Assessment of other literature for apple suggests that the basipetal transport of IAA from scion to root is an important physiological signal regulating root growth. For example, exogenous auxins promoted the initiation of apple roots (Jones and Hatfield, 1976; Delargy and Wright, 1979) and composite apple trees on dwarfing rootstocks generally have smaller root systems than vigorous rootstocks (Hatton et al., 1923; Rogers and Vyvyan, 1934; Colby, 1935; Vyvyan, 1955; Beakbane and Rogers, 1956; Tubbs, 1980; Abod and Webster, 1989). Smaller root systems of dwarfing apple rootstocks typically decrease the size and vigour of the scion (Tubbs, 1980) and similar non-rootstock induced vigour reductions can be imposed for apple scions by root pruning (Ferree, 1992; Ferree et al., 1992) or root restriction (Webster et al., 2000; Atkinson et al., 2000; Webster et al., 2003). However, it is unknown whether the physiological mechanisms modified by root restriction are the same as those regulating rootstock-induced dwarfing of the scion.

The stem tissue of dwarfing apple rootstocks can also reduce the growth of vigorous root systems, possibly by restricting the basipetal transport of IAA. For example, the size of the M.12 root system was greatly reduced when budded with a M.9 scion and the resulting composite tree was much smaller than M.12 budded onto the M.12 rootstock (Vyvyan, 1955). Similarly, budding or grafting a length of dwarfing shoot or interstem between the scion and a vigorous rootstock reduced root and scion growth (Tukey and Brase, 1943), and the dwarfing effect on the scion was increased with increasing length of the interstem used (Parry and Rogers, 1972). In addition, the grafting of bark inserts of M.26 into composite trees of ‘Gravenstein’ on M.111 greatly reduced scion vigour and the dwarfing effect was markedly increased by inverting the bark (Lockard and Schneider, 1981), presumably because reversing tissue polarity greatly decreases the basipetal transport of radio-labelled IAA (Antoszewski et al., 1978).

In summary, dwarfing apple rootstocks may decrease the basipetal transport of IAA within their rootstock stem to the root system. Sub-optimal amounts of IAA transported to the root system of a dwarfing apple rootstock might limit root growth and, consequently, scion vigour. As discussed below, reduced scion vigour on a dwarfing apple rootstock may result from reduced basipetal IAA transport to the root modifying root-produced hormonal signals transported to the scion in the xylem vasculature.

1.7.3.2 Effect of dwarfing apple rootstocks on root to shoot signalling of cytokinin and gibberellin

Lockard and Schneider (1981) hypothesised that reduced basipetal transport of IAA from scion to root decreased root growth and the consequent amount of cytokinin transported to the scion in the xylem sap, thereby limiting extension growth of the scion. This hypothesis is physiologically feasible because auxin is important for the initiation of apple roots (Jones and Hatfield, 1976; Delargy and Wright, 1979). In addition, it has long been known that root tips are important sites of cytokinin biosynthesis, which was convincingly demonstrated in a recent study of Arabidopsis

(Nordstrom et al., 2004). Therefore, decreased basipetal transport of IAA from scion to root of trees growing on dwarfing apple rootstocks may reduce cytokinin biosynthesis by decreasing the number of root tips. Furthermore, inhibition of the basipetal IAA signal may down-regulate cytokinin metabolism by the root directly (Lockard and

Schneider, 1981). However, short-term decapitation studies for other plant species have shown that decreased basipetal transport of IAA from shoot to root increased the concentration of cytokinin transported in the xylem sap (Currie, 1997; Bangerth et al., 2000; Nordstrom et al., 2004). Therefore, decreased basipetal transport of IAA may not reduce cytokinin biosynthesis directly.

Nevertheless, cytokinins are present in the xylem sap of apple trees (Jones, 1973; Young, 1989; Cutting et al., 1991; Skogerbo and Mage, 1992; Tromp and Ovaa, 1994; Kamboj, 1996; Kamboj et al., 1999) suggesting that they are produced by the root and transported to the scion in the xylem vasculature (Jones, 1964, 1967, 1973). In addition, the ‘Fiesta’ scion grafted onto M.9 had lower total concentrations of zeatin plus zeatin riboside in the xylem sap than MM.106 (Kamboj, 1996; Kamboj et al., 1999). Therefore, dwarfing rootstocks do reduce the amount of cytokinin transported from root to scion. Jones (1973) showed that cytokinin stimulated the outgrowth of isolated apple shoots in vitro, and subsequently it was postulated that increased endogenous concentrations of cytokinin within the xylem sap might explain the increased rates of scion growth for composite trees on vigorous rootstocks (Kamboj et al., 1999).

However, it has long been known that exogenous cytokinin does not increase the mean shoot length of primary (Wertheim and Estabrooks, 1994) or secondary shoots (Forshey, 1982; Elfving, 1984, 1985; Cody et al., 1985; Miller and Eldridge, 1986; Popenoe and Barritt, 1988) of young composite apple trees. Furthermore, the cytokinin benzylaminopurine (BAP) applied to young apple scions primarily stimulated axillary buds along the primary shoot to break and form secondary shoots (Williams and Stahly, 1968; Kender and Carpenter, 1972; Forshey, 1982; Elfving, 1985; Miller and Eldridge, 1986; Popenoe and Barritt, 1988; Volz et al., 1994; Wertheim and Estabrooks, 1994). In contrast, in a tree nursery there were trends that M.9 reduced the mean number of secondary shoots that formed on different scion cultivars in their first year of growth (Jaumien et al., 1993; Volz et al., 1994). Similarly, the ‘Worcester Pearmain’ scion on M.9 was smaller than that on M.16 partly because of fewer growing points (Avery, 1969). Therefore, the literature indicates that a decreased supply of cytokinin transported to the scion growing on a dwarfing apple rootstock (Kamboj et al., 1999) may predominantly cause a scion phenotype with fewer growing points.

The inability of exogenous cytokinin to stimulate additional growth of the primary (Wertheim and Estabrooks, 1994) and secondary shoots (Forshey, 1982;Elfving, 1984, 1985; Cody et al., 1985; Miller and Eldridge, 1986; Popenoe and Barritt, 1988) for scions growing on young composite apple trees strongly indicates that other endogenous hormones are necessary to stimulate shoot extension growth of the scion on a dwarfing apple rootstock. A recent study at East Malling reported that gibberellins were virtually undetectable in the xylem sap of scions grown on different size-controlling rootstocks (East Malling, 2005). In addition, the most recent review into the physiological causes of rootstock-induced dwarfing of the scion reported that gibberellins might not be important in the dwarfing mechanism (Webster, 2004).

In contrast, other evidence suggests that gibberellins may indeed be important in rootstock-induced dwarfing of the scion. For example, gibberellins are transported within the xylem sap of apple trees (Jones and Lacey, 1968; Ibrahim and Dana, 1971; Motosugi et al., 1996) indicating that the apple root may synthesise and supply gibberellins to the scion (Jones and Lacey, 1968). Dwarfing compared with vigorous rootstocks decreased endogenous concentrations of gibberellins within the root (Yadava and Lockard, 1977), xylem sap (Ibrahim and Dana, 1971) and leaves or shoots (Yadava and Lockard, 1977; Fontana-Degradi and Visai, 1978). More importantly, exogenous gibberellin(s) stimulated shoot extension growth of apple (Sironval et al., 1962; Marcelle, 1963; Martin, 1967; Robitaille and Carlson, 1971, 1976; Luckwill and Silva, 1979; Tromp, 1982; Steffens et al., 1985; Popenoe and Barritt, 1988). Exogenous GA3 did not increase the rate of node emergence by the apple shoot apical meristem, but increased the proportion of shoots that were growing late in the season (Luckwill and Silva, 1979). In contrast, an important way in which dwarfing apple rootstocks decreased shoot growth compared with vigorous rootstocks was to increase the proportion of shoots that terminated early in the growing season (Swarbrick, 1929; Colby, 1935; Tubbs, 1951; Avery, 1969; Robitaille and Carlson, 1976). Therefore, earlier termination of SAMs for a scion growing on a dwarfing apple rootstock may result from the dwarfing rootstock transporting lower amounts of root-produced gibberellin to the scion in the xylem vasculature.

The literature for apple also indicates that the basipetal transport of IAA from scion to root is important for the growth of shoot apical meristems. For example, the application

of the auxin transport inhibitor ‘TIBA’ to the root/shoot transition region of ‘Antonovka’ apple seedlings reduced the basipetal transport of 14C-IAA and caused the eventual termination of the SAM on the primary shoot (Grochowska et al., 1994). In a similar manner, the bark of the dwarfing apple rootstock metabolised more IAA (Martin and Stahly, 1967; Gur and Samish, 1968) and its stem reduced the basipetal transport of radio-labelled IAA (Soumelidou et al., 1994a; Kamboj et al., 1997), particularly as shoot growth slowed late in the season (Kamboj et al., 1997).

Therefore, shoot/root/shoot signalling mechanisms may exist whereby dwarfing apple rootstocks decrease the basipetal transport of IAA from scion to root that in turn reduces gibberellin synthesis by the root system and its consequent transport to the scion, therefore limiting extension growth of the scion by increasing shoot termination. In addition, reduced basipetal transport of IAA from scion to root may reduce root growth, the biosynthesis of cytokinins and their transport to the scion, thereby causing a scion phenotype with fewer growing points.

1.8 Summary, rationale and thesis objectives

Adoption of high-density orchard systems planted on dwarfing apple rootstocks has potential to greatly improve production efficiency of New Zealand apple orchards. However, there is a strong need to breed new dwarfing apple rootstocks that are resistant to woolly apple aphid, fire blight, phytophthora and SARD. At present, breeding and selection of new apple rootstocks for commercial release is largely empirical and, therefore, time consuming.

Recent research has identified a dwarfing locus (DW1) involved in the dwarfing mechanism of the M.9 rootstock (Rusholme-Pilcher et al., 2008) and a genetic map has been constructed of apple rootstock progeny derived from a cross between M.9 and ‘Robusta 5’ (Celton et al., 2009). These are important scientific advancements because the identification of genetic markers linked to dwarfing gene(s) will enable desirable progeny to be selected from large populations of tree material at a very young age. Efficiency and effectiveness of rootstock breeding will, therefore, be improved. In addition, detailed genetic maps have potential application in further elucidating the genetic control of rootstock-induced dwarfing of the scion.

Despite recent advances in the genetic elucidation of rootstock-induced dwarfing of the scion, the fundamental biological processes that dwarfing gene(s) control are poorly understood, particularly the underlying physiological mechanism(s) that are first modified within the composite apple tree growing on a dwarfing rootstock, and their consequent expression in scion architecture during early tree phenology. Important physiological mechanisms by which dwarfing apple rootstocks decrease scion vigour may involve restricting the endogenous transport of nutrients, water and hormones. The most plausible of these is the modification of shoot/root/shoot signalling of endogenous